Oral squamous cell carcinomas (OSCC) represent the 6th most common cancer in the United States and are associated with low survival and high morbidity. Unfortunately, the 50% survival rate of patients with OSCC has not improved in the last 20 years. Furthermore, aggressive treatment of OSCC is disfiguring and can lead to long-term functional problems. One approach to reducing morbidity and mortality in OSCC patients is to accurately predict which tumors will be aggressive. The current staging system is inadequate in predicting tumor aggressiveness and patient outcome because of the wide variation in tumor behavior. This heterogeneity in tumor behavior is believed to reflect differences in the underlying gene expression profiles of individual cancers. Recently, gene expression profiling has identified tumor aggressiveness signatures and predictors of patient outcome in breast cancer, lung cancer, prostate cancer, etc. An aggressiveness signature and outcome predictor has not been described for OSCC. This R21 proposal will test the hypothesis that gene expression profiling using high-density microarrays will identify a gene signature for OSCC tumor aggressiveness and a molecular predictor for patient outcome. This hypothesis will be explored using three specific aims. 1.) Obtain gene expression profiles using paired normal and OSCC specimens from our Head and Neck Tumor Bank. 2.) Identify a molecular signature for OSCC aggressiveness and a predictor for patient outcome. 3.) Validate the molecular signatures with an independent sample of patients. This study will involve collaborations between investigators with expertise in microarrays, tumor biology, statistics, medicine, pathology, and bioinformatics. The data from this work will serve as preliminary data in R01/R33 applications testing the clinical utility of customized OSCC aggressiveness and patient outcome arrays as well as applications directed at understanding OSCC cancer biology and the development of new OSCC therapies. The unique molecular signature for OSCC tumor aggressiveness and the predictor of patient outcome established from this work has the enormous potential to reduce OSCC patient mortality and morbidity.